CN103677066A - Photovoltaic array maximum power point tracing method of quick self-adapting step length - Google Patents

Abstract

The invention discloses a photovoltaic array maximum power point tracing method of a quick self-adapting step length. The method comprises the steps that a small maximum power point neighbourhood U (Pmax, epsilon) is determined, the step length of the provided self-adapting step length v step is automatically adjusted, and the maximum step length delta v max is kept to be tracked outside the maximum power point neighbourhood U (Pmax, epsilon); in the maximum power point neighbourhood U (Pmax, epsilon), the self-adapting step length v step is further reduced fast, the step length v step at the maximum power point is kept to be zero, tracking precision is guaranteed, and meanwhile the tracking speed of a system is improved.

Description

A kind of photovoltaic array maximum power point tracing method of quick self-adapted step-length

Technical field

The present invention relates to photovoltaic array maximum power point tracing method, especially relate to a kind of photovoltaic array maximum power point tracing method of quick self-adapted step-length.

Background technology

MPPT maximum power point tracking (MPPT) technology is almost a most important part in photovoltaic system.At constant voltage process, disturbance observation (Perturb and Observe, abbreviation P & O), incremental conductance method (Incremental Conductance, be called for short IncCond) etc. on classical way basis, proposed a large amount of improved MPPT methods and be applied.Wherein, for overcoming the contradiction of MPPT maximum power point tracking accuracy and runtime, adaptive step method is arisen at the historic moment.

Recently, A.K.Abdelsalam is at document " High-performance adaptive perturb and observe MPPT technique for photovoltaic-based microgrids " (IEEE Transactions on Power Electronics, 26 (4), 2011) propose to produce self-adaptation disturbance by the difference DELTA P of two continuous power signals of photovoltaic array, he defines adaptive step v _{step_1}=M * | Δ P|.Wherein, M value is the definite constant of experiment, | Δ P| represents the difference of power signal | P _i-P _i-1|.K. ?J.Lee at document " An adaptive maximum power point tracking scheme based on a variable scaling factor for photovoltaic systems " (IEEE transactions on Energy conversion, 27 (4), 2012) by formula

distortion, obtains dP _max=I * dU+U * dI, definition adaptive step v _{step_2}=M * | I * dU+U * dI|.Choosing of these two kinds of adaptive steps is in fact v _step=M * | the distortion of dP/dU|, can guarantee when apart from maximum power point position far away adaptive step v _steplarger; And near maximum power point time, automatically reduce step-length v _step, to eliminate near the vibration causing because of large step-length maximum power point, improved tracking accuracy.But this adaptive step v _stepaffected greatly by photovoltaic property curve P-V rate of curve, when P-V rate of curve is larger, adaptive step v _steplarger; When P-V rate of curve hour, adaptive step v _stepless, this has just limited its tracking velocity.

A.Ahmed in document " A Fast PV Power Tracking Control Algorithm With Reduced Power Mode " (IEEE transactions on Energy conversion, 28 (3), 2013) by formula

distortion, definition

wherein max (| dU/dI|) represent the experiment to produce one group | dU/dI| data are got maximal value.Near (dP/dU>0) in the process of maximum power point, when U increases, | dU/dI| reduces, and | U/I| increases, and causes adaptive step v _stepreduce, until at maximum power point place | dU/dI|=|U/I|, guarantees adaptive step v _step=0.But the method, when solving tracking velocity and this conflict of tracking accuracy, still fails to propose effective strategy.

Disclosed patent of in March, 2013 " a kind of for the quick self-adapted maximum power point tracing method of photovoltaic array step-length ", for the problems referred to above, proposes following strategy: definition adaptive step

V _{step_4}=Δ v _max* ln (| dP/dV|+1)/ln (| dP/dV|+1)+1, meets when P-V rate of curve is larger adaptive step v _steplarger; When P-V rate of curve hour, still can keep relatively large adaptive step v _step, until arrive peak power vertex neighborhood U (P _max, ε) in, just start to reduce rapidly step-length, thereby when guaranteeing tracking accuracy, improved tracking velocity.

So, how by qualitative analysis, to be converted into quantitative test, to determine a less peak power vertex neighborhood U (P _max, ε), become when guaranteeing tracking accuracy, further improve the key of tracking velocity, also become emphasis of the present invention simultaneously.

Summary of the invention

To be solved by this inventionly be to provide a kind ofly when guaranteeing tracking accuracy, further improve the photovoltaic array maximum power point tracing method of the quick self-adapted step-length of tracking velocity.

The present invention solves the problems of the technologies described above taked technical scheme: a kind of photovoltaic array maximum power point tracing method of quick self-adapted step-length, comprises following steps:

1., after system starts, photovoltaic array output voltage and output current are sampled, V (k) expression k (k=0,1,2 ..., the k) output voltage of inferior sampling, I (k) expression k (k=0,1,2 ..., the k) output current of inferior sampling; Wherein, initial voltage V (0) is 80% of open-circuit voltage Vr, i.e. V (0)=Vr * 80%;

Power variation dP=I (k) * V (k)-I (k-1) * V (k-1) while 2. calculating the k time sampling, and dP carries out condition judgment:

A) if meet dP=0, adaptive step v _step=0, can obtain one group of output voltage V (k), output current I (k), making output power P (k)=V (k) * I (k) is maximum power point, realizes the tracking of maximum power point;

B) if meet dP ≠ 0, output voltage variable quantity dV=V (the k)-V (k-1) while calculating the k time sampling, and dV is carried out to condition judgment:

If a) meet dV=0, output current variable quantity dI=I (the k)-I (k-1) while calculating the k time sampling, and dI is carried out to condition judgment:

If i. meet dI=0, step-length v _step=0, can obtain one group of output voltage V (k), output current I (k), making output power P (k)=V (k) * I (k) is maximum power point, realizes the tracking of maximum power point;

If ii. meet dI ≠ 0, return to step 1.; Until meet dI=0, step-length v _step=0, can obtain one group of output voltage V (k), output current I (k), making output power P (k)=V (k) * I (k) is maximum power point, realizes the tracking of maximum power point;

B) if meet dV ≠ 0, calculate x=dP/dV, wherein x represents photovoltaic array P-V rate of curve;

I. x is carried out to condition judgment:

If meet 0≤x<1, calculate adaptive step v _step=Δ v _max* | x-xlnx|;

If meet-1<x<0, calculate adaptive step v _step=Δ v _max* || x|-|x|ln|x||;

If meet x>=1 or x≤-1, calculate adaptive step v _step=Δ v _max;

Δ v wherein _max=M * Vr represents maximum step value; M represents adaptive step correction constant; ||

Represent to ask absolute value operation;

Ii. revise and alternative output voltage values V'(k)=V (k)+v _step, output current is carried out to resampling by I'(k) and alternative I (k);

Iii. as adaptive step v _step, return to step 1. at ≠ 0 o'clock;

As adaptive step v _stepwithin=0 o'clock, can obtain one group of output voltage V ' (k), output current I'(k), make output power P'(k)=V'(k) * I'(k) for maximum power point, realize the tracking of maximum power point.

Compared with prior art, the present invention first calculates dP, avoids arriving maximum power point, but still blindly calculates adaptive step v _stepspecial circumstances occur; DV, dI are calculated, and reply changes dI ≠ 0 because illuminance abrupt variation causes output current, and adaptive step v _step=0 special circumstances, system restart, finds new maximum power point.Main advantage of the present invention is to determine a less peak power vertex neighborhood U (P _max, ε), make (the P at peak power neighborhood of a point U _max, ε) outside, i.e. dP/dV>=1 or dP/dV≤-1 o'clock, the adaptive step v of proposition _step=Δ v _max, keep maximum step delta v _maxmaximum power point is followed the tracks of; At peak power neighborhood of a point U (P _max, ε) in ,-during 1<dP/dV<1, the adaptive step v of proposition _step=Δ v _max* || dP/dV|-|dP/dV|ln|dP/dV||, reaches and reduces rapidly adaptive step, and guarantee the step-length v of maximum power point place _step=0 requirement, makes, when guaranteeing tracking accuracy, to have improved the tracking velocity of system.

Accompanying drawing explanation

Fig. 1 is photovoltaic property curve P-V curve;

Fig. 2 is flowage structure block diagram of the present invention;

Fig. 3 A is matlab emulation adaptive step v _{step_1}=M * | Δ P|,

v _{step_2}=M×|I×dU+U×dI|、

$v_{\mathrm{step}\_3}=M\&times;\left|\frac{|\mathrm{dU}/\mathrm{dI}|-|U/I|}{\max \left(\right|\mathrm{dU}/\mathrm{dI}\left|\right)}\right|,$

v _{step_4}=Δv _max×ln(|dP/dV|+1)/ln(|dP/dV|+1)+1、

$v_{\mathrm{step}\_5}=\left\{\begin{array}{cc}\mathrm{\&Delta;}{v}_{\max }\&times;\left|\right|\mathrm{dP}/\mathrm{dU}|-|\mathrm{dP}/\mathrm{dU}\left|\ln \right|\mathrm{dP}/\mathrm{dU}\left|\right|& ,|\mathrm{dP}/\mathrm{dU}|<1\\ \mathrm{\&Delta;}{v}_{\max }& ,|\mathrm{dP}/\mathrm{dU}|\&GreaterEqual;1\end{array}\right.$

Time curve;

Fig. 3 B is the partial enlarged drawing of Fig. 3 A;

Fig. 4 A is matlab emulation adaptive step v _{step_1}=M * | Δ P|,

v _{step_2}=M×|I×dU+U×dI|、

$v_{\mathrm{step}\_3}=M\&times;\left|\frac{|\mathrm{dU}/\mathrm{dI}|-|U/I|}{\max \left(\right|\mathrm{dU}/\mathrm{dI}\left|\right)}\right|,$

v _{step_4}=Δv _max×ln(|dP/dV|+1)/ln(|dP/dV|+1)+1、

$v_{\mathrm{step}\_5}=\left\{\begin{array}{cc}\mathrm{\&Delta;}{v}_{\max }\&times;\left|\right|\mathrm{dP}/\mathrm{dU}|-|\mathrm{dP}/\mathrm{dU}\left|\ln \right|\mathrm{dP}/\mathrm{dU}\left|\right|& ,|\mathrm{dP}/\mathrm{dU}|<1\\ \mathrm{\&Delta;}{v}_{\max }& ,|\mathrm{dP}/\mathrm{dU}|\&GreaterEqual;1\end{array}\right.$

Time MPPT maximum power point tracking curve;

Fig. 4 B is the partial enlarged drawing of Fig. 4 A.

Embodiment

Below in conjunction with accompanying drawing example, the present invention is described in further detail.

As shown in Figure 2, a kind of photovoltaic array maximum power point tracing method of quick self-adapted step-length, comprises the following steps:

1., after system starts, photovoltaic array output voltage and output current are sampled, V (k) expression k (k=0,1,2 ..., the K) output voltage of inferior sampling, I (k) expression k (k=0,1,2 ..., the K) output current of inferior sampling; Wherein, initial voltage V (0) is 80% of open-circuit voltage Vr, i.e. V (0)=Vr * 80%; For example, adopt typical solar panel Solarex MSX60, open-circuit voltage Vr=21.000V; So, initial voltage V (0)=Vr * 80%=16.800V, initial current I (0)=3.5952A; Sample for the first time, i.e. k=1; V (1)=16.905V, I (1)=3.5760A.

Power variation dP=I (k) * V (k)-I (k-1) * V (k-1) while 2. calculating the k time sampling, and dP carries out condition judgment:

A) if meet dP=0, step-length v _step=0, can obtain one group of output voltage V (k), output current I (k), making output power P (k)=V (k) * I (k) is maximum power point, realizes the tracking of maximum power point;

B) if meet dP ≠ 0, output voltage variable quantity dV=V (the k)-V (k-1) while calculating the k time sampling, and dV is carried out to condition judgment:

If a) meet dV=0, output current variable quantity dI=I (the k)-I (k-1) while calculating the k time sampling, and dI is carried out to condition judgment:

If i. meet dI=0, step-length v _step=0, can obtain one group of output voltage V (k), output current I (k), making output power P (k)=V (k) * I (k) is maximum power point, realizes the tracking of maximum power point;

If ii. meet dI ≠ 0, return to step 1.; Until meet dI=0, step-length v _step=0, can obtain one group of output voltage V (k), output current I (k), making output power P (k)=V (k) * I (k) is maximum power point, realizes the tracking of maximum power point;

B) if meet dV ≠ 0, calculate x=dP/dV, wherein x represents photovoltaic array P-V rate of curve;

I. x is carried out to condition judgment:

If meet 0≤x<1, calculate adaptive step v _step=Δ v _max* | x-xlnx|;

If meet-1<x<0, calculate adaptive step v _step=Δ v _max* || x|-|x|ln|x||;

If meet x>=1 or x≤-1, calculate adaptive step v _step=Δ v _max;

Δ v wherein _max=M * Vr represents maximum step value; M represents adaptive step correction constant; || represent to ask absolute value operation;

Ii. revise and alternative output voltage values V'(k)=V (k)+v _step, output current is carried out to resampling by I'(k) and alternative I (k);

Iii. as adaptive step v _step, return to step 1. at ≠ 0 o'clock;

As adaptive step v _stepwithin=0 o'clock, can obtain one group of output voltage V ' (k), output current I'(k), make output power P'(k)=V'(k) * I'(k) for maximum power point, realize the tracking of maximum power point.

According to above step, while sampling for the first time, i.e. k=1; Power variation dP=I (1) * V (1)-I (0) * V (0)=0.053W, output voltage variable quantity dV=V (1)-V (0)=0.105V.Obvious dP ≠ 0, dV ≠ 0, calculates x=dP/dV=0.505>=0 and x<1, so, adaptive step v _step=Δ v _max* | x-xlnx|=0.089.Revise and alternative output voltage values V'(1)=V (1)+0.100=16.994V, output current is carried out to resampling by I'(k) the alternative I (k) of=3.5584A.Obviously, adaptive step v _step, return to step 1., until as adaptive step v at ≠ 0 o'clock _stepwithin=0 o'clock, can obtain one group of output voltage V ' (k), output current I'(k), make output power P'(k)=V'(k) * I'(k) for maximum power point, realize the tracking of maximum power point.

Feasibility and the validity of MPPT maximum power point tracking of the present invention (MPPT) method can further illustrate by following simulation result.

Table 1

Table 1 is under same test platform (the theoretical maximum power point power of test platform 60.4727W, maximum power point voltage 17.02V), the emulated data of different MPPT methods.Table 1 shows that, under identical test platform, the inventive method is with adaptive step v _{step_1}and v _{step_2}compare, tracking accuracy is improved, and approaches theoretical maximum power point; Aspect tracking velocity, increase, wherein compared with adaptive step v _{step_4}improve 34.8%.Adopt the adaptive step simulation curve of the inventive method and other four kinds of methods more as shown in Figure 3, and MPPT maximum power point tracking simulation curve more as shown in Figure 4.In Fig. 3, peak power vertex neighborhood is expressed as to U (17.02,0.27).Only analyze photovoltaic array P-V family curve ascent stage, adaptive step v _{step_3}meet the position v far away apart from maximum power point _{step_3}larger; And apart from the nearer position of maximum power point and even to maximum power point place, adaptive step v _{step_3}monotone decreasing is zero condition, but v _{step_3}there is not definite maximum step delta v _max, in the selection of constant M, there is the contradiction between tracking velocity and tracking accuracy.V _{step_2}, and v _{step_1}similar, at the position V=14.00V far away apart from maximum power point, just start to reduce, compare v _{step_4}and v _{step_5}, need the time more of a specified duration to reach maximum power point.At the nearer position V ∈ (14.00V, 17.29V) of maximum power point, v _{step_4}start progressively to reduce, and v _{step_5}still keeping larger, stable step-length, until in peak power vertex neighborhood U (17.02,0.27), v _{step_5}just start to reduce rapidly, and at the step-length v of maximum power point place _{step_5}=0, obtained higher tracking velocity and tracking accuracy.Fig. 4 shows under identical test platform, and the tracking velocity of the inventive method is better than other four kinds of methods.Table 2 is five kinds of MPPT methods tracking data when following the tracks of maximum power point in Fig. 4.Associative list 2 and Fig. 4 are known, v _{step_4}need 5 steps to complete the tracking of 60.4727W; v _{step_2}in the 9th step, complete the tracking of 60.4679W, tracking accuracy is higher, has 0.008% error with theoretical maximum power point; v _{step_3}in the 9th step, complete the tracking of 60.4234W, tracking accuracy is not high, has 0.082% error with theoretical maximum power point; And step-length function v in this paper _{step_5}only need 3 steps to complete the tracking of 60.4727W, show that the inventive method, when guaranteeing tracking accuracy, has improved the tracking velocity of system.

Table 2

Power points	v step_1	v step_2	v step_3	v step_4	v step_5
						1	26.5763	32.4762	60.1774	60.2008	60.4529
2	48.1562	53.5172	60.3041	60.3235	60.4719
						3	59.8171	57.8430	60.4092	60.4155	60.4727
4	56.8518	55.2887	60.4511	60.4608	?
						5	59.2934	58.4491	60.4653	60.4727	?
6	60.2408	60.1690	60.4701	?	?
						7	60.4035	60.4538	60.4718	?	?
8	60.4214	60.4674	60.4727	?	?
						9	60.4234	60.4679	?	?	?

Claims (1)

1. a photovoltaic array maximum power point tracing method for quick self-adapted step-length, is characterized in that comprising following steps:

1., after system starts, photovoltaic array output voltage and output current are sampled, V (k) expression k (k=0,1,2 ..., the k) output voltage of inferior sampling, I (k) expression k (k=0,1,2 ..., the k) output current of inferior sampling; Wherein, initial voltage V (0) is 80% of open-circuit voltage Vr, i.e. V (0)=Vr * 80%;

Power variation dP=I (k) * V (k)-I (k-1) * V (k-1) while 2. calculating the k time sampling, and dP carries out condition judgment:

A) if meet dP=0, step-length v _step=0, can obtain one group of output voltage V (k), output current I (k), making output power P (k)=V (k) * I (k) is maximum power point, realizes the tracking of maximum power point;

B) if meet dP ≠ 0, output voltage variable quantity dV=V (the k)-V (k-1) while calculating the k time sampling, and dV is carried out to condition judgment:

If a) meet dV=0, output current variable quantity dI=I (the k)-I (k-1) while calculating the k time sampling, and dI is carried out to condition judgment:

If i. meet dI=0, step-length v _step=0, can obtain one group of output voltage V (k), output current I (k), making output power P (k)=V (k) * I (k) is maximum power point, realizes the tracking of maximum power point;

If ii. meet dI ≠ 0, return to step 1.; Until meet dI=0, step-length v _step=0, can obtain one group of output voltage V (k), output current I (k), making output power P (k)=V (k) * I (k) is maximum power point, realizes the tracking of maximum power point;

B) if meet dV ≠ 0, calculate x=dP/dV, wherein x represents photovoltaic array P-V rate of curve;

I. x is carried out to condition judgment:

If meet 0≤x<1, calculate adaptive step v _step=Δ v _max* | x-xlnx|;

If meet-1<x<0, calculate adaptive step v _step=Δ v _max* || x|-|x|ln|x||;

If meet x>=1 or x≤-1, calculate adaptive step v _step=Δ v _max;

Δ v wherein _max=M * Vr represents maximum step value; M represents adaptive step correction constant; || represent to ask absolute value operation;

Ii. revise and alternative output voltage values V'(k)=V (k)+v _step, output current is carried out to resampling by I'(k) and alternative I (k);

Iii. as adaptive step v _step, return to step 1. at ≠ 0 o'clock;

As adaptive step v _stepwithin=0 o'clock, can obtain one group of output voltage V ' (k), output current I'(k), make output power P'(k)=V'(k) * I'(k) for maximum power point, realize the tracking of maximum power point.

Images